Poor Man's OCR Post-Correction

Hammarström, Harald; Virk, Shafqat Mumtaz; Forsberg, Markus

doi:10.1145/3078081.3078107

Cited by 7 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In total, 34 teams registered to the competition, for a final number of 5 submissions. Note that we received an extra answer from a Swedish team that applied their word-based method [6] but couldn't get more than 10-20% on the training dataset, so they decided to not submit their results. The following section gives a brief description of the 5 submitted methods.…”

Section: Submitted Methodsmentioning

confidence: 99%

“…The digitized documents come from different collections available, among others, in national libraries or universities. The corresponding GT comes from initiatives such as HIMANIS 4 , IMPACT 5 , IMPRESSO 6 , Open data of National Library of Finland 7 , GT4HistOCR [4] and RECEIPT [5].…”

Section: B Datasetmentioning

confidence: 99%

“…http://www.himanis.org5 https://www.digitisation.eu6 https://impresso-project.ch 7 https://digi.kansalliskirjasto.fi/opendata…”

unclassified

See 2 more Smart Citations

ICDAR 2019 Competition on Post-OCR Text Correction

Rigaud

Doucet

Coustaty

et al. 2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

View full text Add to dashboard Cite

This paper describes the second round of the ICDAR 2019 competition on post-OCR text correction and presents the different methods submitted by the participants. OCR has been an active research field for over the past 30 years but results are still imperfect, especially for historical documents. The purpose of this competition is to compare and evaluate automatic approaches for correcting (denoising) OCR-ed texts. The present challenge consists of two tasks: 1) error detection and 2) error correction. An original dataset of 22M OCR-ed symbols along with an aligned ground truth was provided to the participants with 80% of the dataset dedicated to training and 20% to evaluation. Different sources were aggregated and contain newspapers, historical printed documents as well as manuscripts and shopping receipts, covering 10 European languages (Bulgarian, Czech, Dutch, English, Finish, French, German, Polish, Spanish and Slovak). Five teams submitted results, the error detection scores vary from 41 to 95% and the best error correction improvement is 44%. This competition, which counted 34 registrations, illustrates the strong interest of the community to improve OCR output, which is a key issue to any digitization process involving textual data.

show abstract

Section: Submitted Methodsmentioning

confidence: 99%

Section: B Datasetmentioning

confidence: 99%

See 1 more Smart Citation

ICDAR 2019 Competition on Post-OCR Text Correction

Rigaud

Doucet

Coustaty

et al. 2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

View full text Add to dashboard Cite

show abstract

“…H. Hammarström et al propose a language-independent Post-correction system, without supervision or human intervention to improve OCR results in historical documents [22]. The idea of the method is based on the fact that two words are variants of each other if their distribution similarity exceeds what is expected by chance of their similarity of form.…”

Section: Methods Similaritymentioning

confidence: 99%

“…22: Results of the substitution method considering the analysis of all the characters with certainty restriction.…”

mentioning

confidence: 99%

Improving text recognition accuracy using syntax-based techniques

Miranda¹

View full text Add to dashboard Cite

First of all, I thank God and Saint Ignatius of Loyola for giving me this opportunity in my life to improve personally and professionally.I am thankful to my parents Sergio and Doris for their advice, unconditional support, patience, love and always being with me. Moreover, I thank my siblings Carola, Ignacio, and Camila, and my aunt Candy for always been there for me and being my best friends. To my grandmother Eulogia for taking care of my family and my grandparents Nicanor and María Esther, my angels who always take care of me.Especially, I would like to thank Prof. Guido and co-adviser Prof. Marcio for their advice, guidance, experience, and knowledge which enabled me to finish this stage of my life.I thank all my friends that I made during these two years, especially those from the Computer Systems Laboratory (LSC), for sharing their knowledge. I would also like to thank my boyfriend for helping me at this stage of my life and always being there when I need it. And especially to the IC secretary, for helping me with the processes at UNICAMP.

show abstract

Inventory and Content Separation in Grammatical Descriptions of Languages of the World

Hammarström

2021

Linking Theory and Practice of Digital Libraries

View full text Add to dashboard Cite

Poor Man's OCR Post-Correction

Cited by 7 publications

References 0 publications

ICDAR 2019 Competition on Post-OCR Text Correction

ICDAR 2019 Competition on Post-OCR Text Correction

Improving text recognition accuracy using syntax-based techniques

Inventory and Content Separation in Grammatical Descriptions of Languages of the World

Contact Info

Product

Resources

About